自适应容量自回归视觉跟踪
Adaptive Capacity Autoregressive Visual Tracking
把扩散模型当作时序难度预测器，用来动态切换高/低容量推理模式，为视觉跟踪提供了一种「按需计算」的自回归框架；该思路可迁移到任何需要在线调节推理开销的时序视觉任务。
原文依据 A diffusion-based difficulty estimator anticipates the stability of upcoming segments, guiding a controller to switch between an accurate (high-capacity) and an efficient (low-capacity) mode
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 ARTrack-AC，这是自回归跟踪范式的新进展，引入自适应容量推理以同时实现时序一致性与动态效率。现有自回归跟踪器以固定的推理容量顺序预测目标状态，无法适应真实视频中波动的时序难度。ARTrack-AC 通过赋予跟踪器随时间调节其推理容量的能力来解决这一局限。一个基于扩散的难度估计器预测即将到来片段的稳定性，引导控制器在精确（高容量）与高效（低容量）模式之间切换，同时保持自回归一致性。这种系统级自回归将传统序列建模从「预测什么」扩展到「如何预测」，形成一种自我调节的跟踪过程，使推理开销与时序复杂度相匹配。尽管方法简单，ARTrack-AC 在主要基准上取得了最优的精度-速度权衡——在 LaSOT 上达到 66.7% AUC，在 LaSOText 上达到 47.5% AUC——运行速度比其前代快 2.9 倍。
英文原文摘要 1133 字符
We present ARTrack-AC, a new step in the autoregressive tracking paradigm that introduces adaptive capacity inference to achieve both temporal consistency and dynamic efficiency. While existing autoregressive trackers predict object states sequentially with fixed inference capacity, they fail to accommodate the fluctuating temporal difficulty of real videos. ARTrack-AC addresses this limitation by equipping the tracker with the ability to modulate its inference capacity over time. A diffusion-based difficulty estimator anticipates the stability of upcoming segments, guiding a controller to switch between an accurate (high-capacity) and an efficient (low-capacity) mode while maintaining autoregressive consistency. This system-level autoregression extends conventional sequence modeling beyond "what to predict" toward "how to predict," forming a self-regulated tracking process that aligns inference cost with temporal complexity. Despite its simplicity, ARTrack-AC achieves state-of-the-art accuracy-speed trade-off on major benchmarks--66.7% AUC on LaSOT and 47.5% AUC on LaSOText--running 2.9xfaster than its predecessor.